UniDS: A Unified Dialogue System for Chit-Chat and Task-oriented Dialogues. Anthology ID: 2022.dialdoc-1.2 Volume: Proceedings of the Second DialDoc Workshop on Document-grounded Dialogue and Conversational Question Answering Month: May Year: 2022 Address: Dublin, Ireland Venue: dialdoc SIG: Publisher: Association for Computational Linguistics Note: Pages: 13–22 Language: URL: DOI: 10.18653/v1/2022.dialdoc-1.2 Bibkey: zhao-etal-2022-unids Cite (ACL): Xinyan Zhao, Bin He, Yasheng Wang, Yitong Li, Fei Mi, Yajiao Liu, Xin Jiang, Qun Liu, and Huanhuan Chen. More importantly, UniDS achieves better robustness than pure dialogue systems and satisfactory switch ability between two types of dialogues. Experimental results demonstrate that the proposed UniDS works comparably well as the state-of-the-art chit-chat dialogue systems and task-oriented dialogue systems. UniDS does not need to adding extra parameters to existing chit-chat dialogue systems. Besides, we propose a two-stage training method to train UniDS based on the unified dialogue data schema. In particular, we design a unified dialogue data schema, compatible for both chit-chat and task-oriented dialogues. To this end, we propose a unified dialogue system (UniDS) with the two aforementioned skills. To achieve more natural interaction with humans, dialogue systems need to be capable of both chatting and accomplishing tasks. However, these two systems are often tackled separately in current methods. Abstract With the advances in deep learning, tremendous progress has been made with chit-chat dialogue systems and task-oriented dialogue systems.
0 Comments
Leave a Reply. |